Improved Digital Image Correlation for Light Source Variation

The Newton-Rapshon(N-R) algorithm based on sum of squared differences function used in digital image correlation(DIC) may get unreliable results in specimen deformation measurement because it does not consider the light source variation. Thus, a modified N-R algorithm based on sum of squared differences that considers the effect of light source variation to the measurement results has been presented in this paper. This proposed algorithm has been programmed into Windows application program by VC software and showed easy operation and accurate measurements. The algorithm has been validated by comparing the computer generated speckle images. The results indicate that the proposed algorithm will be more effective than former algorithm when the surrounding light source changes. At the end of this paper, the relationship between the proposed method and zero-normalized sum of squared differences (ZNSSD) function that is always used to remove the light source change effect has been analyzed; the characteristics of both methods have been presented.